Live Sensibly (with alcohol), 07-27-2004: Moderate Drinking: By the Numbers

July 27, 2004

• Single Entry Page • Add a Comment

Moderate Drinking: By the Numbers

How much is too much? Two drinks? Five? Ten?

Oprah Winfrey asked that question on her show, but the answers she got seemed muddy to me.

It sounds like a simple question, but to be fair, the most accurate answers aren’t going to be monolithic or apply universally. Even the experts’ answers aren’t all the same. Let’s poke around some of the available answers from U.S. dietary guidelines, a study about harm thresholds, and recommendations in Britain and other countries.

Note that:

  • The definition of a drink is on this page.
  • Dual thresholds (one per day for women, two for men, for example) are used because of the physiological differences in the ways men and women metabolize and are impacted by alcohol.
  • The companion to this page is: Moderate Drinking: Beyond the Numbers.

Below: Frameworks for moderate drinking: (a) One/Two drinks daily; (b) U.K.: Three/Four units daily; (c) Three/Four daily with 9/14 weekly; and, (d) Higher Levels, and then a Summary.

One/Two Drinks Daily

The NIAAA cites the Dietary Guidelines for Americans (233K PDF), also known as the Food Pyramid, as the source for its numeric definition of moderate drinking. The Food Pyramid was developed jointly by the U.S. Department of Agriculture and the U.S. Department of Health and Human Services. From page 40 of the guidelines:

If adults choose to drink alcoholic beverages, they should consume them only in moderation.


Moderation is defined as no more than one drink per day for women and no more than two drinks per day for men.

Current evidence suggests that moderate drinking is associated with a lower risk for coronary heart disease in some individuals.

The guidelines also list risks of drinking more than the moderate amount, as well as the conditions under which folks are better off abstaining.

The NIAAA’s 1992 Alcohol Alert, "Moderate Drinking" discounts the usefulness of quantifying moderate drinking, and makes the following points:

  • Moderate drinking (i.e., that which does not generally cause problems) is often confused with — but is distinct from — social drinking (patterns that are generally accepted in the society in which they occur).
  • The accuracy of numeric definitions of moderation is tempered by the wide range of effects a given dose of alcohol may have on people of different sizes, genders, and drinking histories.
  • Evidence suggests that moderate drinking may have psychological and cardiovascular benefits.
  • Negative effects of moderate drinking may include higher incidence of stroke, vehicle crashes, medication interactions, birth defects, and future alcohol abuse and/or dependence.

U.K.: Three/Four Units Daily

In the U.K., an interdepartmental group of officials was convened in 1994 to review medical and scientific evidence on the long-term effects of drinking alcohol. Its results were published in the December, 1995 report "Sensible Drinking". (If you’re interested in a detailed review of the evidence, follow the link to the PDF version of the report — it is a thorough accounting of global scientific evidence as of 1995.)

Drink units are measured in increments of 8 grams of pure alcohol in the U.K., compared to 14 grams of alcohol per drink in the U.S. The Sensible Drinking report’s recommendations included:

MEN: Regular consumption of between 3 and 4 units a day by men of all ages will not accrue significant health risk.

WOMEN: Regular consumption of between 2 and 3 units a day by women of all ages will not accrue any significant health risk.

The International Center for Alcohol Policies (ICAP) analyzed the distinctions between the U.S. one/two drink and the U.K. three/four unit recommendations in this report (360K PDF):

The alcohol levels deemed safe for women by UK standards are approximately 70% higher than the levels deemed safe in the US. The difference in levels deemed safe for men, on the other hand, is only 17%.

Three/Four Drinks Daily, 9/14 Weekly

Moderation Management names slightly higher daily limits than the NIAAA, while its recommended weekly limits are in sync with the NIAAA definition.

MM Limits:

  • Strictly obey local laws regarding drinking and driving.
  • Do not drink in situations that would endanger yourself or others.
  • Do not drink every day. MM suggests that you abstain from drinking alcohol at least 3 or 4 days per week.
  • Women who drink more than 3 drinks on any day, and more than 9 drinks per week, may be drinking at harmful levels.
  • Men who drink more than 4 drinks on any day, and more than 14 drinks per week, may be drinking at harmful levels.

Notes: (1) BAC info, BAC charts; (2) Standard drink definition

The MM limits are based on research by Martha Sanchez-Craig, DA Wilkinson and R Davila: Empirically based guidelines for moderate drinking.

Dr. Reid Hester, a researcher and clinician who supports MM and SMART (short bio here), also offered this definition of moderate drinking in 2000:

It’s no more than 2 or 3 standard drinks per drinking episode, no more than 9 drinks per week for women and 12-14 for men. Also, moderate drinking means limiting how fast you drink and, as a result, keeping your blood alcohol concentration (BAC) below .045-.055.

Evidence of Low Risk at Higher Levels

Dr. Stanton Peele and Dr. Archie Brodsky published a paper on the psychological benefits of moderate drinking. In it, they noted the challenges of defining moderate drinking.

Within the U.S., …the level of drinking linked with lowest mortality has sometimes been measured to be substantially higher than [the one/two drink daily limit] (Greenfield et al., 1999). …Rehm and Bondy (1998) reported, “Heavier-drinking cohorts tend to display their minimum risk at relatively higher levels of alcohol intake than cohorts with lower alcohol consumption” for which there was “no satisfactory explanation” (p. 223).

Cohorts? I tell ya, the number of cohorts lurking in my daily vocabulary is … ahhh … well … none. To me, the basic point is that some folks drinking above the one/two daily threshold have not experienced a proportional increase in risk.

Related to its three/four unit recommendation, the U.K. Sensible Drinking report noted (emphasis added):

The problem drinker

(10.7) Our recommendations are for the individual drinker in the normal drinking population. They are not framed particularly to influence clinical treatment of problem drinkers or indeed their recognition. We wish to move away from a culture of advice on consumption levels which has been interpreted by some as categorising all those who drink above the currently recommended levels as heavy or problem drinkers when, clearly, the vast majority of them are not.

In The Natural History of Alcoholism, Dr. George Vaillant spoke of a study which followed 204 men from their sophomore year of college — 1940 — through 1980:

The men in the College sample have reported their alcohol use relatively accurately every 2 years for 40 years. Between the ages of 40 and 60, several men regularly recorded drinking six ounces (four drinks) of whiskey a day — or more than a gallon a month — for more than 20 years without problems. However, no man in our College sample reported drinking over five drinks a day without reporting unwanted symptoms and concern over his capacity to control his drinking.

So, in this limited sample (men aged 40-60 who had attended an elite college), a 20-year pattern of four drinks per day — equal to MM’s daily limit, but double the MM and USDA weekly limits — produced no measurable consequences.

Finally, ICAP (the International Center For Alcohol Policies) analyzed moderate drinking recommendations around the globe, and published the results in this report supplement (31K PDF). A table on the second page of the report notes that recommended moderate levels in Austria, Canada, Denmark, New Zealand and Romania are in ranges roughly similar to those in the U.S. and U.K, while those in Australia, Italy, and Japan are higher.


The Food Pyramid gives us a structure within which to understand what our bodies typically need to be healthy. It is based on some of the best data we’ve got, and yet it’s not the only data we’ve got. It doesn’t dictate the only valid path we can take — some of us have rearranged it substantially to follow low-carb or vegan diets, for example — nor does it make value judgments if we choose to eat more pizza or fewer vegetables than recommended.

(In round numbers, the guidelines in the official food pyramid document use the word “can” about 40 times, “should” 20 times — 3 of them related to drinking — and an unconditional “must” only once, when noting that vegans must supplement with B12.)

That seems like a good model, to me, for understanding its one/two daily drink guideline. If we want to compete as athletes, survive the rigors of medical residency, qualify as astronauts, or other endeavors requiring top-notch physical conditioning, we will need to heed the pyramid, including the drink limits. Those of us who seek reasonably good (but less than Olympian) health and functioning should understand the risks of stretching the pyramid’s boundaries, but there is some wiggle-room built in: Consuming a little more pizza, few less vegetables, or a bit more wine is generally not going to turn an otherwise contented life sour and short.

Balance, of course, is the key. The best way to create it is by examining the experts’ recommendations in the context of our values and goals so that we can make well-informed choices.

(Companion page: Moderate Drinking: Beyond the Numbers.)

  • posted by Bose
  • created 27-Jul-2004
  • last updated 22-Aug-2004

Post a comment

(or nickname)
(will not be displayed)
  Save my details locally: